Ask an Acoustic Engineer (me)

[Jscoyne2]

At first glance, I was thinking you meant "box stuffing" before my brain kicked in - and since my brain is wired for car audio rather than real life, before I realized what you really meant, I said "Yes, box stuffing IS sort of like what @ckirocz28 is talking about!"

It's funny you raise this topic though, here.

In my house, I'm currently looking for sound absorbing materials, simply because I have a finished basement with a drop ceiling, and the cold air return goes over the middle of the space down there... so any sound, TV, stereo, video games, anything - only has a thin sheet metal cold air return preventing the sound from being piped up to the first floor via the cold air return. So in my case, I'm looking for a material that will BLOCK sound. And in the home space, there are sound-absorbing boards, sound-absorbing acoustic foam (like anechoic chambers or recording studios), and I've recently been told by a professional commercial-space drywall buddy of mine to look into mineral wool - sold pretty cheaply in the insulation aisles of home centers.
Those materials are designed to absorb sound, so it doesn't pass through them.

(box stuffing does that also - but now makes me wonder why we use polyfill rather than mineral wool, since I believe mineral wool has better properties)

On the other hand, if my cold air return was buzzing and vibrating every time my furnace kicked on, that would be annoying sound caused by the panels themselves. Same if my basement stereo system caused panels to buzz every time certain songs were played loud. in that case, I'd be looking for damping materials - like Dynamat or Second Skin (or whatever the kids use these days) to stop the panels from vibrating. You'll notice if you close a cheap car's door, it makes a higher pitched "empty can" sound than the dull "thud" of closing a luxury car door. That's mainly the effect of damping materials lowering the resonance of the doors. If I apply that material to those panels that vibrate - I'll (hopefully!) stop them from vibrating and resonating.

And a third type of acoustic absorption might actually be what you are talking about... In a car we've got a terrible combination of hard, reflective glass and plastic surfaces - and plush, soft, foam-filled seats. The glass causes some reflections, and the seats cause some of the sound to be absorbed. Both, in combination wreak havoc on our actual frequency response at our seating position. I hope you aren't asking for a simple (or even complex) way to predict these effects in any accurate way?

More of links to white papers on any research that may have been done on sound absorption in a vehicle. I've looked through quite a few paper and there is A SHITLOAD of really cool tech that's been out for a few years that is helping auto manufacturers find modes and resonance within a digital model of their vehicle. They were using ray tracing for high frequency and some tech i forgot the name of for low freq, but they knew exactly where cancellation would be based on T/a and driver location. Pretty interesting stuff. I know all of.. me and 1 other person who has attempted heavy handed sound absorption in a car. Its hard to quantify a difference with and without, but i also don't how to measure for reverb and such. Which is what im trying to educate myself on now

[geolemon]

More of links to white papers on any research that may have been done on sound absorption in a vehicle. I've looked through quite a few paper and there is A SHITLOAD of really cool tech that's been out for a few years that is helping auto manufacturers find modes and resonance within a digital model of their vehicle. They were using ray tracing for high frequency and some tech i forgot the name of for low freq, but they knew exactly where cancellation would be based on T/a and driver location. Pretty interesting stuff. I know all of.. me and 1 other person who has attempted heavy handed sound absorption in a car. Its hard to quantify a difference with and without, but i also don't how to measure for reverb and such. Which is what im trying to educate myself on now

Apologies - I just intended to describe and differentiate two very different types of "sound absorption" in my post - both deal with vibrational energy, but two very different applications:
1) materials designed to block sound energy (eg. acoustic foam)
2) materials designed to dampen vibrations carried through a surface (eg. damping material)

As far as what materials work best for either category - I'm sure there are studies in each category, anywhere from formal white papers to casual comparisons of "which product is best" (which themselves often vary in quality). But to get you started - this initial differentiation alone could get you un-stuck, if that's where you were stuck in your research.
It is frequently confusing - these are two very different categories of application, that all too frequently get commingled in discussions. That was my point, essentially.

I know this little bit, that frequencies are a factor, for both:
For sound transmission - bass is harder to block than treble. That's true no matter the material. I'm sure simple density plays a role there, but at least at higher frequencies there are materials that can absorb sound energy, converting it into heat (which is actually what the previously mentioned box stuffing does).
For vibration damping - the whole goal IS to convert higher frequency vibrations into lower frequency, adding mass is usually the technique, but there's also materials that try to actively absorb - and especially for those, I'm sure frequency becomes a factor.

As far as measuring reverb or echo... that is a great question.
I'm thinking you could actually use a 2 channel oscilloscope - with a frequency generator going to one channel (I suppose that would be optional, if you only have a 1 channel, like I do - but there are inexpensive PC-based (USB) oscilloscopes out there that do two channel) and also to your amplifiers, and a microphone connected to the other. In that way, you could send impulse bursts into your amplifier and speaker/speakers in your listening space... so trace 1 would show you the reference signal - perfectly clean with no echo, including where it ended - and your trace 2 would show you the measured response at the position of the microphone, including echo. You would have to factor in sound transmission time (speed of sound) to properly align the two signals, but that sounds like a pretty scientific approach to me - but that's just my intuition. It would be time consuming, because I think you'd want to burst individual frequencies, working your way across the frequency spectrum, to find resonances and echoes. I'm not aware of a reference rig or specific tool, other than the visual comparison that would offer.

Even in the highest levels of car audio, most people take an approach like this:
"Hey, something is vibrating in my car when I'm listening really loud."
"Let's take a look - if it's something that's loose or vibrating or rubbing, we can secure it down. We can also use sound damping material to stop panels from vibrating. Can you put something on that makes it do what you want to stop?"

[Jscoyne2]

Apologies - I just intended to describe and differentiate two very different types of "sound absorption" in my post - both deal with vibrational energy, but two very different applications:
1) materials designed to block sound energy (eg. acoustic foam)
2) materials designed to dampen vibrations carried through a surface (eg. damping material)

As far as what materials work best for either category - I'm sure there are studies in each category, anywhere from formal white papers to casual comparisons of "which product is best" (which themselves often vary in quality). But to get you started - this initial differentiation alone could get you un-stuck, if that's where you were stuck in your research.
It is frequently confusing - these are two very different categories of application, that all too frequently get commingled in discussions. That was my point, essentially.

I know this little bit, that frequencies are a factor, for both:
For sound transmission - bass is harder to block than treble. That's true no matter the material. I'm sure simple density plays a role there, but at least at higher frequencies there are materials that can absorb sound energy, converting it into heat (which is actually what the previously mentioned box stuffing does).
For vibration damping - the whole goal IS to convert higher frequency vibrations into lower frequency, adding mass is usually the technique, but there's also materials that try to actively absorb - and especially for those, I'm sure frequency becomes a factor.

As far as measuring reverb or echo... that is a great question.
I'm thinking you could actually use a 2 channel oscilloscope - with a frequency generator going to one channel (I suppose that would be optional, if you only have a 1 channel, like I do - but there are inexpensive PC-based (USB) oscilloscopes out there that do two channel) and also to your amplifiers, and a microphone connected to the other. In that way, you could send impulse bursts into your amplifier and speaker/speakers in your listening space... so trace 1 would show you the reference signal - perfectly clean with no echo, including where it ended - and your trace 2 would show you the measured response at the position of the microphone, including echo. You would have to factor in sound transmission time (speed of sound) to properly align the two signals, but that sounds like a pretty scientific approach to me - but that's just my intuition. It would be time consuming, because I think you'd want to burst individual frequencies, working your way across the frequency spectrum, to find resonances and echoes. I'm not aware of a reference rig or specific tool, other than the visual comparison that would offer.

Even in the highest levels of car audio, most people take an approach like this:
"Hey, something is vibrating in my car when I'm listening really loud."
"Let's take a look - if it's something that's loose or vibrating or rubbing, we can secure it down. We can also use sound damping material to stop panels from vibrating. Can you put something on that makes it do what you want to stop?"

So in my last car i completely rebuilt both doors to have 8" thick acoustic foam all the way through. Basically each door was a sound trap. Under both seats was as much foam as i could fit. I tore out the entire center console and in the gap was a bunch of acoustic foam. The dash had 1.5in melamine foam across the entire dash. But..i also went from a 3 way(10in midbass, 6.5 mid/ 1in tweet) to a hlcd/8in, during that entire foaming of my car. So i really didnt get to compare a before and after.

After some experimenting and a lot of reading. It seems like absorption in a car is pretty tricky because of so many reflections. I tried to get some answers from people on diyma and messaging some smarter people like the Magic Bus guy and some MIT guys but mostly the answers i got were, "No idea. Try it. Cars are weird"..which isn't much help. I think it may have even but Jazzi that tried it but he(or someone) put a whole ass 4x4x4ft block insulation in his hatchback and found it was a pretty damn effective bass trap. I believe that if you have drivers in your pillars or dash, that gutting your dash and either building one deep and long acoustic trap/ or one long deep diffusion panel of varying depths, could do wonders to making the windshield disappear. Might also make it way worse or the drivers more locatable or the stage more blurry.

It just seems after years of building my own system and being on Diyma. I've seen huge rebuilds of every part of a car in the search for sound quality but almost no one tries to build a bass trap or reflection ansorbers into their car and i just dont get why..and im not smart enough to know how to really test certain things out objectively and with proof.

I was just looking for maybe something technical where people had tried. Maybe people tried in the 90s hayday and found it wasn't worth it. With as much money as finals people put into their cars, if absorption worked. They'd probably do it right?

This is what John Whitledge had to say, "as you know cars are hostile acoustical environments.

''your approach with acoustic foam is well-intentioned but flawed. the LAST thing a car needs is broadband acoustical absorption. furthermore, foam only works at frequencies for mid- to high- frequencies. using too much foam will suck the life out of the music andcreate unbalanced reverberation spectra. lastly, you would be taking a “shotgun” approach to acoustical treatments thinking that too much of a good thing is not possible. there’s no point in expending effort where it is not needed - but, how do you know where effort IS needed, and what is the best solution? science must be employed.

you’ve done well by damping the sheet metal body.

what you need are mostly DIFFUSIVE surfaces to scatter sound uniformly, and minimal absorption only exactly where you need it, like at points of first reflection (think ray tracing, see my part 6 article on my website).

just like in the magic bus you really only need bass traps like Helmholtz absorbers.

it also helps to decouple (isolate) the vibrations of your subwoofer and mid basses from the body of the car.

i see no value in rear fill speakers, or loudspeaker placement in "the cone of confusion”. doing so creates more problems than is solves. simplicity is key to good sound"

The reverb spectra is what caught my eye. RoomEq does a RT60 Decay graph which i think is the same thing? Idk i have to do some reading. Pushing into territory im ignorant about here.

[dgage]

If you have a SINGLE home subwoofer, play it and then walk around the room. Somewhere in the middle you'll find reduced bass output. In all or some of the corners you should notice more bass and usually boomy bass. These are the bass waves interacting with the room. To mitigate this, it is suggested to have multiple subwoofers, up to 4 before diminishing returns, to even out the bass in a room. To cut down on the boominess that is common in the corners, bass traps are built.

The car isn't a rectangle and doesn't have as pronounced and large corners where you might use a bass trap so I'm not sure that part of the equation would cross over cleanly to the auto environment.

But regarding your use of foam for absorption, in a home you also want some absorption and diffraction. All of one or the other makes a very poor audio environment. Since we know a car has plenty or too much diffraction due to the many hard and different angled surfaces, it is a good idea to add as much absorption as you can. Under the seat and dash sound like great places to try to stop audio waves from going under there and bouncing around. A more cushioned headliner would be a huge improvement for absorption though it would only affect the highest frequencies since most headliners are pretty thin.

I think the other issue you're running into with a car is size and shape, at least with studies. With a room, you can do some tests in a room and the results would translate to many other rooms. With a car, you'd have to test different sizes and shapes as sedans react differently than hatchbacks and huge SUVS. And while Harmon might be able to financially benefit from experiments in rooms since many studios and people can leverage that knowledge for their room, which would drive sales, the same can't necessarily be said of a car. Car audio is a small enough market and shrinking that it doesn't make sense to invest tons of money to experiment on a car's acoustics. Or the companies that spend the money like Bose, aren't generally going to share that research, unfortunately.

[Justin Zazzi]

I'd like to educate myself more on the idea of absorption and materials and how they're rated. Perhaps how acoustic absorption works in a car.

I was reading about the Magic Bus awhile back and while he is.. eccentric..he was talking about having the acoustic energy or decay in a room be equal across the spectrum. I don't recall exactly what it was.

Im looking for reading material that doesn't take a degree to understand. Really anything on absorption that anyone here would recommend.

I think you're talking about reverberation time which is usually measured with "RT60" or "RT30". Room EQ Wizard can calculate this for you. Essentially it means how long does it take for the room to get quiet again after the music stops. This is, like all audio things, frequency-dependent. The goal that Jon was chasing is to have a specific reverberation time for all frequencies to better mimic the behavior of a performance hall or a place where music is performed ... instead of sounding like a car which is relatively dead.


image borrowed from Jon's website here: https://www.magicbus.biz/measurements.html

It will be hard to find reading about absorption in car interiors perhaps because it is not studied much. There are some hints in one of my favorite books Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms by Floyd Toole. That book goes into great detail about reverberations, reflections, diffractions, and all the other things that a room can contribute to a sonic experience. If you haven't read it yet, I love recommending it.

Other than that, you will find a ton of information about absorption in books on building recording studios, performance spaces, and especially performance halls or opera houses.

Here is a post in my build log about absorption:
https://www.diymobileaudio.com/threa...3/post-1706290

Here's another one about reflections:
https://www.diymobileaudio.com/threa...3/post-1719939

What would you like to learn about absorption?

[Justin Zazzi]

but i also don't how to measure for reverb and such. Which is what im trying to educate myself on now

If in doubt, room eq wizard can probabbly do it! look on the rt60 tab:



Also, you can see a similar effect in the waterfall plots. I think these are easier to visualize for some people? The waterfall below and the RT60 plot above are the same data set. Notice the lower RT time above in the treble matches a faster decay of the treble in the waterfall below.

After some experimenting and a lot of reading. It seems like absorption in a car is pretty tricky because of so many reflections. I tried to get some answers from people on diyma and messaging some smarter people like the Magic Bus guy and some MIT guys but mostly the answers i got were, "No idea. Try it. Cars are weird"..which isn't much help. I think it may have even but Jazzi that tried it but he(or someone) put a whole ass 4x4x4ft block insulation in his hatchback and found it was a pretty damn effective bass trap. I believe that if you have drivers in your pillars or dash, that gutting your dash and either building one deep and long acoustic trap/ or one long deep diffusion panel of varying depths, could do wonders to making the windshield disappear. Might also make it way worse or the drivers more locatable or the stage more blurry.

Lol yeah that was me

Here's the pots in my build log where I tried to change the RT60 time of my car by adding a bass trap, inspired by Jon's work on his bus:
https://www.diymobileaudio.com/threa...3/post-1803758

The reason I found sound treatments to be rare in a car is they are impractical. Bass traps require a phenomenal amount of space to work correctly. Jon's bus has over a dozen traps hidden in the panels, roof, even the sliding doors have traps in them. My bass trap consumed my entire cargo area. Reflection enhancements like the binary amplitude diffusors that Jon installed are essentially a wall with holes in it which is impossible to put into a daily-driver.

Since we know a car has plenty or too much diffraction due to the many hard and different angled surfaces, it is a good idea to add as much absorption as you can. Under the seat and dash sound like great places to try to stop audio waves from going under there and bouncing around. A more cushioned headliner would be a huge improvement for absorption though it would only affect the highest frequencies since most headliners are pretty thin.

Reverberation time is all about energy vs time. Imagine the distance sound can travel from four reflections in a typical living room: ten feet from stereo to the wall, 20 feet to the other wall, 15 feet to another wall, 15 feet to another wall .... that's 60 feet of absorption-free travel off four reflections. Now imagine Can you plot 60 feet of absorption-free reflections in your car? It's pretty tough!

Cars do have a lot of reflective surfaces but the reflective surfaces frequently are interrupted by soft objects like passengers, upholstery, seats. So while there are many surfaces to reflect off of, there is more-than-usual absorption material between one hard surface and another in a car. This is why reverberation time in a car is small because the sound is absorbed quicker over time. Counter-intuitive isn't it?

This is why Jon tried to *increase* the amount of reflective surfaces in his bus by adding the binary amplitude diffusor panels behind the front seats. This is also the heart of his quote above that Jscoyne2 found.


image borrowed from Jon's website here: https://www.magicbus.biz/photos.html

[Justin Zazzi]

Also look for "part 6" of the magazine series below. Jon might be passionate, but his methodology is pretty solid:

https://www.magicbus.biz/authored-articles.html

[ckirocz28]

Like a bandpass?
Attachment 11708

No. Like this.

[ckirocz28]

I think like Holmz said - a bandpass box would get you close, although it's more the speaker between two boxes.
What you are asking is "what would the effect be, if you built a big sealed box with a little vented box just sitting inside it - no speaker attached to the vented box, just a box with a port in it" right?
Let's think through it to see if we can get close-
I'd believe sure enough the sub in the sealed box would cause pressure fluctuations in the sealed box... which would in turn act on the air in the vent, which would move in resonance with whatever frequency it was tuned to given that box... and that would follow the same rules as vented boxes, as far as the vent resonating in phase or out of phase with the cone...
That WOULD have the excursion limiting effects on cone motion (I believe) above the tuning frequency, since the port air motion in this case would purely be adding to or decreasing the effective pressure in the enclosure - same as a vented box, but not [significantly, anyway] contributing to the sound outside the box, since the vented box is trapped inside the sealed box.
Also similar to a vented box, I would think that below the vent tuning frequency, the vent-effect collapses, and now you'd re-capture the air inside the vented box, for use by the sealed box.
Same goes if you used a passive radiator, if my thinking above is all correct.

So let's assume we stuck a 1 cu.ft. empty box with a vent (or passive radiator) in it tuned to 40hz (but no sub or sub hole) inside of a sealed box that was large enough to give us a 2 cu.ft. sealed volume even with that vented box inside it...
if the air inside box one is Vb1 and the air inside box two is Vb2, I'd predict (but I don't know for a fact):

• Below 35-40hz, the sub would behave as if it were in a 3 cu.ft. box. So if you modeled up a 3 cu.ft. box, and looked at 40hz and down - that's what you'd get.
• Right near 40hz, the sub and vent are as close to in-resonance as it gets - which would in this case cause the vent to add pressure inside the sealed box causing a sharp reduction in excursion, causing a dip in frequency response.
• Above 40hz, the sub and vent go progressively out-of-phase, so that destructive interference would fade away, until you get to whatever frequency the vent wouldn't have any impact any more... above that frequency, we're behaving like we're in a 2 cu.ft. sealed box... which at those upper frequencies is going to be pretty close to the 3 cu.ft. response anyway.

So - overall, I think if you modeled your sub in a 3 cu.ft. box and imagined a decent dip in the frequency response around 40hz (I'm guessing about 6dB), and then rising back up to the -0dB level again progressively by about an octave higher (1 octave is also a guess) as you move further up the frequency spectrum.

But that's just me trying to think through it - I can't say I know that is what would happen. I don't believe there's any software that models this, to know for sure.

Do you have a need for a sealed box with a dip in the response? I could only see this as being really beneficial for subs stuffed in sealed boxes that are too small and have peaky responses to iron out - but in those cases of course you don't have room to stuff a vented box inside, or you'd have enough air space to begin with.

You actually completely understood what I was thinking of! I have no need of this, I just thought it was an interesting thought experiment, if not an actual physical experiment.

[Jscoyne2]

I think you're talking about reverberation time which is usually measured with "RT60" or "RT30". Room EQ Wizard can calculate this for you. Essentially it means how long does it take for the room to get quiet again after the music stops. This is, like all audio things, frequency-dependent. The goal that Jon was chasing is to have a specific reverberation time for all frequencies to better mimic the behavior of a performance hall or a place where music is performed ... instead of sounding like a car which is relatively dead.


image borrowed from Jon's website here: https://www.magicbus.biz/measurements.html

It will be hard to find reading about absorption in car interiors perhaps because it is not studied much. There are some hints in one of my favorite books Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms by Floyd Toole. That book goes into great detail about reverberations, reflections, diffractions, and all the other things that a room can contribute to a sonic experience. If you haven't read it yet, I love recommending it.

Other than that, you will find a ton of information about absorption in books on building recording studios, performance spaces, and especially performance halls or opera houses.

Here is a post in my build log about absorption:
https://www.diymobileaudio.com/threa...3/post-1706290

Here's another one about reflections:
https://www.diymobileaudio.com/threa...3/post-1719939

What would you like to learn about absorption?

So thats a super interesting graph. If im reading that right. We have very little reverberation in a car until basically below the Schrodinger frequency( i think im using that right). If everything below 100hz is already in the green, Whats the point of a bass trap? Is it too make the bass frequencies more flat and around the 0.1, 0.2 level?

If his idea is to mimic the reverb of a large listening area, how does one increase reverb in a car?

and what is reverb?


also straight from Rew.
"RT60 is a measure of how long sound takes to decay by 60 dB in a space that has a diffuse soundfield, meaning a room large enough that reflections from the source reach the mic from all directions at the same level. Domestic rooms are usually too small to have anything approaching a diffuse field at low frequencies as their behaviour in that region is dominated by modal resonances. As a result RT60 is typically not meaningful in such rooms below a few hundred Hz. Use the waterfall, spectrogram and decay plots to examine the decay of low frequencies in domestically-sized rooms."

So maybe rt60 wont work for cars?

My understanding of waterfall graphs is basically that sound is sticking around longer at certain frequencies. What that means exactly or how that sounds. I dont actually know.